{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/cross-domain-self-supervised-multi-task","title":"Cross-Domain Self-supervised Multi-task Feature Learning using Synthetic Imagery","arxiv_id":"1711.09082","date":"2017-11-24","proceeding":"CVPR 2018 6","authors":["Zhongzheng Ren","Yong Jae Lee"],"abstract":"In human learning, it is common to use multiple sources of information\njointly. However, most existing feature learning approaches learn from only a\nsingle task. In this paper, we propose a novel multi-task deep network to learn\ngeneralizable high-level visual representations. Since multi-task learning\nrequires annotations for multiple properties of the same training instance, we\nlook to synthetic images to train our network. To overcome the domain\ndifference between real and synthetic data, we employ an unsupervised feature\nspace domain adaptation method based on adversarial learning. Given an input\nsynthetic RGB image, our network simultaneously predicts its surface normal,\ndepth, and instance contour, while also minimizing the feature space domain\ndifferences between real and synthetic data. Through extensive experiments, we\ndemonstrate that our network learns more transferable representations compared\nto single-task baselines. Our learned representation produces state-of-the-art\ntransfer learning results on PASCAL VOC 2007 classification and 2012 detection.","url_abs":"http://arxiv.org/abs/1711.09082v1","url_pdf":"http://arxiv.org/pdf/1711.09082v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"cross-domain-self-supervised-multi-task","repo_url":"https://github.com/jason718/game-feature-learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.09082","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}